Correlates of the Attitude Toward Life-Sustaining Treatment

Abstract

The goal of the present study was to examine factors associated with the attitude toward life-sustaining treatment among older adults in South Korea. Focus was given to sociodemographic characteristics (age, gender, education, financial status, and religious affiliation), family-related variables (presence of a spouse, children, and a living parent [or in-law]), and health-related variables (chronic conditions, functional disability, and self-rated health). Using data from 6,895 participants (aged 65−97) in the 2014 Korean National Elderly Survey, logistic regression model was tested. The proportion of the sample that endorsed a strong disagreement with the provision of life-sustaining treatment was 53.2%. The increased odds of having a strong disagreement was found in those with younger age (<75), higher education, a living parent (or in-law), and more favorable ratings of health. Findings shed light on programs and services on end-of-life care planning for older adults in South Korea.

Keywords

life-sustaining treatment older Koreans life course end of life Asian elders

Introduction

Accompanied by population aging and advancement in medical technologies, end-of-life care has become a critical issue (Institute of Medicine, 2014; Morhaim & Pollack, 2013; Pew Research Center, 2006). Globally, an increasing number of individuals and families face difficult decisions on whether to withdraw or withhold life-sustaining medical treatment for their incapacitated loved ones. The end-of-life decision-making is of particular concern in South Korea where the rate of population aging is rapid. The proportion of individuals aged 65 and older was 13.5% in 2015, and the rate is estimated to exceed 20% by 2026 (Korean National Statistical Office, 2011). The concern is exacerbated by the fact that medical decision-making for the elderly in Korea is normally assumed by family members, primarily spouse and eldest son (Jo & An, 2013; Kim, 2011; Ko, Nelson-Becker, Park, & Shin, 2013). Beyond the role of family, legislative and regulatory policy on end-of-life care in South Korea has yet to be established (Heo, 2013; Lee, Goo, & Cho, 2016; Shin, Lee, Cho, Yoo, Kim, & Yoo, 2016).

Over the past decade, there has been a growing public interest in patient autonomy and “death with dignity” in South Korea (Heo, 2013; Ko et al., 2013; Lee et al., 2016; Shin et al., 2016). However, most studies on end-of-life care are based on small samples of terminally ill patients, caregivers, or medical professionals, with limited generalizability of findings to the overall Korean population. One exception is a study of over 1,000 members of the general public in South Korea (aged 20−70), where a majority (90%) were shown to support withdrawal of futile life-sustaining treatment (LST; Yun et al., 2011). This rate was similar to those reported by the samples of cancer patients, family caregivers, and oncologists also included in the study (Yun et al., 2011). The study also showed that a higher income was associated with the support for the withdrawal of LST.

Targeting the older population in South Korea, the goals of the present study were to explore their attitude toward LST and to examine its associated factors. Moving beyond sociodemographic characteristics, attention was also paid to family- and health-related variables. Considering the emphasis on familism and filial piety deeply embedded in Korean culture (Sung, 2001), it was expected that the availability of family members, such as spouse, adult children, and even parents, would shape older individuals’ attitude toward end-of-life care. As a result of the increased longevity and co-survival of the generations, for example, it is not uncommon for older Koreans to have a living parent (or in-law). From a life cycle perspective, the presence of a living parent in old age would impact the way in which older individuals perceive their age, social position and role, health, and death (Angel & Settersten, 2011). Studies also suggest that older individuals’ physical health status determines their social status and capacity to be engaged in decision-making processes (Kim, 2011; Schafer, 2011).

To pursue our investigation, we utilized the 2014 Korean National Elderly Survey. This nationally representative sample of older adults in South Korea includes questions relating to LST, offering a unique opportunity to explore the level and associated factors of the attitudes toward LST. The current study was aimed to examine how the attitudes toward LST would be associated with sociodemographic characteristics (age, gender, education, financial status, and religious affiliation), family-related variables (presence of a spouse, children, and a living parent [or in-law]), and health-related variables (chronic conditions, functional disability, and self-rated health).

Methods

Data Set

Data came from the 2014 Korean National Elderly Survey conducted by the Korea Ministry of Health and Welfare. The Korean National Elderly Survey is a nationwide study aimed to explore the status of living arrangement, family and social life, and health and well-being of the elderly population in South Korea (Korea Institute for Health and Social Affairs, 2014). Surveys were conducted with community-dwelling adults aged 65 and older who were selected by a stratified two-stage cluster sampling method. In-person interviews were conducted by trained interviewers between June and September 2014. The original data include a total of 10,451 participants. More information on sampling procedures and methods of the overall study is available elsewhere (Korea Institute for Health and Social Affairs, 2014). The final sample size used in the present analyses was 6,895, after removing participants whose interview was done by a proxy, whose cognition was not intact (Mini-Mental State Examination score <24), and whose response to the question on the attitude toward LST was missing.

Measures

The outcome variable was operationalized using a single question: “What do you think about providing medical treatment when a patient is unconscious or there is no hope of a cure (i.e., LST)?” Response format included strongly agree, agree, neither agree nor disagree, disagree, or strongly disagree. Our preliminary analysis showed that the responses were skewed. As with previous studies (Kim, 2011; Yun et al., 2011), a majority of the sample endorsed some levels of disagreement with the provision of futile LST. To increase the sensitivity of the measure, we dichotomized the responses into 1 = strongly disagree and 0 = otherwise and treated the endorsement of a strong disagreement with LST as an outcome variable. The focus on the extreme response is also legitimized by the fact that participants’ subjective report to sensitive survey items such as end-of-life issues can be biased in the socially desirable direction (Krumpal, 2013).

Scores of all predictor variables were also dichotomized for easy identification and interpretation of group characteristics. Although dichotomization reduces score variance, it allows calculation of odds ratio and efficiently identifies target groups that need to be prioritized in interventions. Cut points were selected based on conventional knowledge, related literature, and descriptive characteristics of the sample.

Sociodemographic information included age (0 = 65−74, 1 = 75 and older), gender (0 = male, 1 = female), education (0 = <high school graduation, 1 = ≥high school graduation), perceived financial status (0 = <adequate, 1 = ≥adequate), and religious affiliation (0 = no, 1 = yes).

Family-related variables were measured with items dealing with whether the respondent had a spouse, children, and one or more living parents (or in-laws). Each response was coded as yes (1) or no (0).

Chronic conditions, functional disability, and self-rated health were used as indicators of physical health status. The total number of chronic conditions was assessed with a list of nine diseases and conditions common in older populations (heart disease, high blood pressure, liver disease, kidney disease, digestive disease, stroke, diabetes, cancer, and arthritis). Functional status was assessed with a list of nine daily activities (eating, dressing, walking, transferring, bathing, toileting, preparing meals, managing medication, and handling transportation); participants indicated whether they needed help with performing each activity. The total numbers of both chronic conditions and functional disability were recoded into a binary variable (0 = none, 1 = one or more conditions/disability). Self-rated health was assessed by asking participants to evaluate their overall health status on a 5-point scale and responses were then dichotomized (0 = excellent/very good/good, 1 = fair/poor).

Analytic Strategy

After assessing frequency distributions, Spearman’s rank-order correlations were used to examine the associations among study variables. Logistic regression of the attitude toward LST was estimated with (a) sociodemographic characteristics, (b) family-related variables, and (c) health-related variables. Due to the complexity of the sampling frame, weighed data were used, and all statistical analyses were conducted using Stata version 13 (Stata Corp, College Station, TX).

Results

Descriptive Characteristics of the Sample

Table 1 presents the descriptive characteristics of the sample. Age ranged from 65 to 97, with an average of 73.2 (SD = 6.14). About 38% of the participants were 75 and older, and more than half (58.5%) were women. About 40% had at least high school education, and less than a half (48.4%) perceived their financial status as less than adequate. Over 31% were not religiously affiliated. The proportion of the sample that had a spouse, children, and a living parent (or in-law) was 62.8%, 97.7%, and 10.6%, respectively. Approximately 80% reported at least one chronic condition, over 10% had limitation in performing at least one activity in the list, and more than 65% rated that their health was either “fair” or “poor.” As anticipated, the attitude toward LST was skewed, about 90% of the sample reporting some levels of disagreement with the provision of futile LST. Our dichotomization focuses on the 53.2% that endorsed a strong disagreement.

Table 1.

Description of the Sample and Study Variables.

	%
Sociodemographic variable
Age
65−74	62.0
≥ 75	38.0
Gender
Male	41.5
Female	58.5
Education
<High school graduation	60.7
≥High school graduation	39.3
Perceived financial status
<Adequate	48.4
≥Adequate	51.6
Religious affiliation
No	31.2
Yes	68.8
Family-related variable
Presence of a spouse
No	37.2
Yes	62.8
Presence of children
No	2.3
Yes	97.7
Presence of a living parent (in-law)
No	89.4
Yes	10.6
Health-related variables
Chronic conditions
None	19.8
One or more	80.2
Functional disability
None	89.4
One or more	10.6
Self-rated health
Excellent/very good/good	34.7
Fair/poor	65.3
Attitudes toward life-sustaining treatment
Strongly agree	0.5
Agree	3.5
Neither agree nor disagree	6.3
Disagree	36.6
Strongly disagree^a	53.2

Note. N = 6,895.

Due to the skewness of the data, the variable was dichotomized into 1 = strongly disagree and 0 = otherwise.

Correlations Among Study Variables

Spearman’s rank-order correlations among study variables were tested (not shown in tabular format). Strong disagreement with LST was associated with age-group (r_s = −.03, p < .05), education (r_s = .03, p < .05), presence of a living parent (in-law) (r_s = .04, p < .01), chronic condition (r_s = −.04, p < .01), and self-rated health (r_s = −.07, p < .001). Individuals with younger age (<75), higher education, a living parent (or in-law), no chronic condition, and more favorable ratings of health were likely to have a strong disagreement with the provision of LST at the end of life.

Logistic regression model of a strong disagreement with LST

Table 2 summarized results of the logistic regression analysis. As for sociodemographic variables, younger age and a higher level of education were associated with increased odds of having a strong disagreement with LST. Individuals who had a living parent (or in-law) showed 1.30 times higher odds of reporting a strong disagreement with LST compared to their counterparts without a living parent (or in-law). Finally, older individuals with fair or poor ratings of health had reduced odds of reporting a strong disagreement with LST.

Table 2.

Logistic Regression Model of a Strong Disagreement With Life-Sustaining Treatment.

	Odds Ratio [95% Confidence Interval]
Sociodemographic variable
Age (reference 65−74)
≥75	0.89* [0.81, 0.98]
Gender (reference male)
Female	1.08 [0.98, 1.19]
Education (reference < High school graduation)
≥High school graduation	1.14* [1.03, 1.26]
Perceived financial status (reference < adequate)
≥adequate	0.94 [0.89, 1.01]
Religious affiliation (reference no)
Yes	1.01 [0.91, 1.12]
Family-related variable
Presence of a spouse (reference no)
Yes	0.93 [0.83, 1.04]
Presence of children (reference no)
Yes	0.78 [0.56, 1.07]
Presence of a living parent (in-law) (reference no)
Yes	1.30** [1.11, 1.52]
Health-related variable
Chronic conditions (reference none)
One or more	0.92 [0.81, 1.04]
Functional disability (reference none)
One or more	0.90 [0.79, 1.07]
Self-rated health (reference excellent/very good/good)
Fair/poor	0.75*** [0.67, 0.83]

p < .05. **p < .01. ***p < .001.

Discussion

Responding to the growing interest in end-of-life care in South Korea (Heo, 2013; Ko et al., 2013; Lee et al., 2016; Shin et al., 2016), the present study explored factors associated with older Koreans’ attitude toward life-sustaining medical treatment. Using a nationally representative sample of older adults (N = 6,895), the predictive role of sociodemographic, family-related, and health-related factors was examined, and findings shed light on programs and services for end-of-life care planning.

It is notable that a majority of the respondents (90%) reported some levels of disagreement with the provision of LST when the benefits were in doubt. The rate is similar to those reported in previous studies with Koreans (Kim, 2011; Yun et al., 2011) and calls attention to a potential response bias. Due to the sensitive nature of the topic, a plausible and culturally informed hypothesis is that participants’ responses are biased in the direction that is assumed to be socially acceptable and desirable (Krumpal, 2013). This might be particularly true for Korean older adults whose culture is based on a collectivism that values not being a burden to others or society as a virtue (Ko et al., 2013; Shin et al., 2016). It is likely that some participants endorsed a disagreement with LST to be in line with the cultural sentiment that “the elderly shall die before being a burden.” To improve the sensitivity of the measure, the present study focused on the 53.2% of the sample that reported a strong disagreement with LST and explored factors associated with their endorsement.

Logistic regression model identified younger age (<75), higher education, presence of a living parent (in-law), and more favorable ratings of health as significant predictors of the strong disagreement with LST. Relative youth, higher educational attainment, and physical health represent personal resources, and older individuals with such resources are likely to have an active engagement in medical decision-making processes and to have a strong opinion on their care preferences (Kim, 2011; Yun et al., 2011).

One of the unique characteristics of the study was its consideration of the presence of a living parent (or in-law) as a potential factor. Over 10% of the present sample had one or more living parents (or in-laws), and their presence was found to be a significant predictor of their attitude toward LST. The co-survival of the aged generations seems to influence the way in which older individuals think about and plan on their own end of care. Considering that adult children assume primary responsibility for elder care in the Korean society, those who are supporting aging parents in their old age may have a particular desire not to be a burden to their children. Beyond age-group or cohort differences, the presence of a living parent (or in-law) plays an important role in contextualizing older individuals’ attitudes toward LST.

An alternative, albeit not mutually exclusive explanation is that those with a greater psychological distance from the need for LST would be most likely to devalue LST. In other words, people who are younger, more educated, in better health, and with still-living parents might not view LST as something relevant, and might be more likely to respond in terms of social desirability pressures. Here it is relevant to point out that the existence of living parents means that an older adult is still a “child” in the eyes of parents and perhaps themselves. In contrast, for someone who is older, less educated, less healthy, and with deceased parents, LST may be more of an immediate reality.

Contrasting to the general disagreement with LST observed in the present study and others (Kim, 2011; Yun et al., 2011), only a small proportion of the Korean population has advance directives. According to the report from the Korean Ministry of Health and Welfare (Ministry of Health and Welfare, 2011), the completion rate of any form of advance directives in South Korea is only 4.7%. The overall findings suggest a gap between individuals’ beliefs and behaviors relating to end-of-life care and call attention to the need for psychoeducational interventions.

Some limitations to the present study need to be noted. Foremost concern is the aforementioned response bias in measuring the attitude toward LST. Although our analyses were performed on the recognition of such bias, caution should be exercised in interpreting findings. Also, family-related variables only considered the presence of a spouse, children, and a living parent (or in-law). Future studies need to consider the qualitative aspects such as support exchange pattern and relationship satisfaction. Finally, the selection of the predictor variable should be expanded by including other potential factors such as the experience of significant other’s death, personal sense of control, religiosity, spirituality, and cultural beliefs.

Despite the limitations, the present study addresses the important end-of-life issue and provides implications for programs and services for older adults in South Korea. Our findings suggest that intervention efforts should include advocating and empowering of older individuals who lack personal resources (e.g., those who are older, less educated, and unhealthy). Interventions should address multiple components including helping older individuals acquire knowledge about end-of-life care planning, build autonomy to express personal choices, facilitate communication with family members on end-of-life care, and learn about the procedures for the completion of advance directives.

Footnotes

Acknowledgments

The authors would like to thank Dr. David A. Chiriboga and Dr. Nan Sook Park for their comments on an earlier version of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2015S1A3A2046745).

Author Biographies

Yuri Jang is a professor in the School of Social Work at the University of Texas at Austin. Before joining UT Austin in 2012, she was a faculty member at the University of South Florida where she had received her doctoral and postdoctoral training in Aging Studies. Her research focuses on positive adaptations in aging, health disparities, and minority health and service utilization. She is currently working on projects on health and healthcare navigation of older ethnic immigrants with limited English proficiency with funding from the National Institute on Aging and City of Austin.

Soo Young Kim is a professor in the Department of Social Welfare at Kyungsung University in South Korea. She received her doctoral degree in Social Welfare from Busan National University. Her research focuses on aging in place and sense of community for the elderly. She is currently leading a project funded by the Social Sciences Korea (SSK) from the National Research Foundation of Korea.

Sujie Chang received her PhD in Social Psychology from Tokyo University. She is currently an associate professor of Social Welfare at Kyungsung University in South Korea. Her research focuses on ways to enhance individual adjustment in later life. Her recent and forthcoming publications address age-friendly community and its influence on the health and well-being of older adults.

References

Angel, J. L., & Settersten, R. A. (2011). Sociology of aging in the decade ahead. In Handbook of sociology of aging (pp. 661–672). New York, NY: Springer.

Heo

D. S.

(2013) Life-sustaining medical treatment for terminal patients in Korea. Journal of Korean Medical Science 28: 1–3. doi: 10.3346/jkms.2013.28.1.1.

Institute of Medicine (2014) Dying in America: Improving quality and honoring individual preferences near the end of life, Washington, DC: National Academies Press.

(2013) Factors affecting shared decision making at end of life in Korean adults. Holistic Nursing Practice 27(6): 329–335. doi: 10.1097/HNP.0b013e3182a72c34.

Kim

S. H.

(2011) Factors influencing preferences of Korean people toward advance directives. Nursing Ethics 18(4): 505–513. doi: 10.1177/0969733011408045.

Nelson-Becker

Park

Shin

(2013) End-of-life decision making in older Korean adults: Concerns, preferences, and expectations. Educational Gerontology 39: 71–81. doi: 10.1080/03601277.2012.682947.

Korea Institute for Health and Social Affairs (2014) 2014 Korean national elderly survey: Policy report 2014–61, Sejong: Korea Institute for Health and Social Affairs.

Korean National Statistical Office (2011) Population projections for Korea: 2010–2060. Based on the 2010 population and housing census, Daejeon: Korea National Statistical Office.

Krumpal

(2013) Determinants of social desirability bias in sensitive surveys: A literature review. Quality & Quantity 47: 2025–2047. doi: 10.1007/s11135-011-9640-9.

10.

Lee

J. E.

Goo

A. J.

Cho

B. L.

(2016) The current status of end-of-life care in Korea and legislation of Well-Dying Act. Journal of Korean Geriatric Society 20(2): 65–70. doi: 10.4235/jkgs.2016.20.2.65.

11.

Ministry of Health and Welfare (2011) Report of the life-sharing perception survey (Report No. 11-1352000-000522-01), Sejong: Ministry of Health and Welfare.

12.

Morhaim

D. K.

Pollack

K. M.

(2013) End-of-life care issues: A personal, economic, public policy, and public health crisis. American Journal of Public Health 1039(6): e8–e10. doi: 10.2105/AJPH.2013.301316.

13.

Pew Research Center (2006) More Americans discussing and planning end-of-life treatment. News Release. 1–44. . Retrieved from http://www.people-press.org/files/legacy-pdf/266.pdf.

14.

Schafer

M. H.

(2011) Health and network centrality in a continuing care retirement community. Journal of Gerontology Series B: Psychological Sciences and Social Sciences 66: 795–803. doi: 10.1093/geronb/gbr112.

15.

Shin

D. W.

Lee

J. E.

Cho

Yoo

S. H.

Kim

Yoo

J. H.

(2016) End-of-life communication in Korean older adults: With focus on advance care planning and advance directives. Geriatrics Gerontology International 16: 407–415. doi: 10.1111/ggi.12603.

16.

Sung

(2001) Family support for the elderly in Korea. Journal of Aging & Social Policy 12(4): 65–79. doi: 10.1300/J031v12n04_04.

17.

Yun

Y. H.

Han

K. H.

Park

B. W.

Cho

C.-H.

Kim

Lee

D. H.

(2011) Attitudes of cancer patients, family caregivers, oncologists and members of the general public toward critical interventions at the end of life of terminally ill patients. Canadian Medical Association Journal 183(10): E673–E679. doi: 10.1503/cmaj.110020